Reconstruction loss hyperparameter

[nordinbelkacemi]

https://github.com/divelab/DIG/blob/dig-stable/dig/fairgraph/method/Graphair/graphair.py#L153

In the code the feature loss is multiplied by beta, but according to the paper, it should be lambda. Is this is a bug?

[hongyiling]

Thanks for pointing out. That's a bug.